audio inductor idea/question

Probably. Toroidal inductors are largely self-shielding, and are used for that reason in many RF filter circuits. Similarly, toroid power transformers are often preferred for their low flux leakage, and thus their reduced tendency to induce hum in nearby low-level audio circuits. I believe that the self-shielding characteristic remains even if you use a core with a very low permeability, such as wood or air... I've seen one homebrew HF radio design in which the author recommended winding toroidal RF inductors using rubber or fiber plumbing washers as a form!

I imagine that you'll need a multiple-layer winding to get enough inductance to be useful at audio frequencies, with a low-permeability core. Winding such toroids, with the relatively thick and stiff wire one would want in order to achieve low series resistance, is likely to be a *major* pain in the thumbs. Toroid power transformers are wound using a special two-part snap-on bobbon - perhaps you can construct something of the sort?

I guess the real question is whether the amount of flux leakage, and thus the amount of cross-coupling between the inductors (in a crossover, I assume?) is great enough to significantly affect the circuit's overall frequency response. The traditional solution is to use cylindrical (solenoid) inductors, and mount them in an orthogonal arrangement (e.g. 90 degree angles between their major axes) in order to minimize the amount of actual cross-coupling. Is that not a sufficient solution for your design?

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Yes this is for crossovers.

Seems like you know Audio guys. Any new fangled approach to get closer to perfection is the goal. The mounting orthoganal is OK until your number of inductors exceeds 3. Then you run out of angles in small enclosures with out resorting to some advanced geomerty.

I would be nice if a indictor could me made was self shielding (without the dreaded high perm material around it) so the corssover doesn't need to resort to fancy geometreis to minimize coupling in a high inductor count crossover.

They can park that next to their $100 1uF Musi-Cap. LOL

Seems to me that an air-core toroid will have considerable external field leakage. It's the hugh-u core that concentrates the flux inside the windings.

John

Real men cross-over-filter _before_ the power amplifiers ;-)

...Jim Thompson

--
| James E.Thompson, P.E.                           |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
         America: Land of the Free, Because of the Brave

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

I had a stereo system once that did this, with separate outputs for the woofer/mid-range and tweeters, but the approach never seemed to catch on. Presumably with "component" systems the problem was that you didn't know what speaker would be attached to being able to adjust the crossover points would have required more additional analog circuitry than the manufacturer wanted to spend money on... whereas these days if you already have a DSP in the box it'd be easy to adjust them with pretty much zero extra cost.

I would think leakage would be close to zero if the winding was very uniform, and dramatically better than other options even if the winding was fairly random. A high-mu core will increase the internal flux and inductance, but won't change much outside the windings. But I wouldn't want to try to DIY crossover toroids.

Tony

and

high-mu

outside the

Suppose a toroidal inductor is lying flat, like a donut on a table, with some current flowing through it. And you're a hungry ant on that table, nearby but outside. The loop of wire nearest you whacks you with some field. The matching loop on the far side whacks you with the opposite field, but it's a lot farther away, so it doesn't cancel the field of the nearer loop. If you treat each loop of the winding as a ring, the ant sees the linear superposition of all the 1-turn fields. None of the flux is especially confined to the interior of the toroid.

Far-field, the cancellation will be better than it would be for a simple solenoid.

I think, for a given amount of wire [1], you get more inductance from a compact solenoid than you get from a toroid. For a compact solenoid, flux from any one turn passes mostly through all the other turns; for the toroid, that's not the case; indeed, flux from some turns passes

*backwards* through other turns.

Without a core to concentrate flux, all sorts of things change.

John

[1] I think somebody figured out the maximum-inductance topology. I think it's a squarish solenoid.

.....snip!.....

It's called a Brooks coil:

, The URL gives you the shape, a round coil, the cross-section of the coil is square (each side is "A", and the radius of the winding is

3A/2. This design is optimized to give you the most inductance for a length of wire, assuming an air core. A large number of crossover coils I've seen use a very similiar physical structure. The shape is inobtrusive, and is pretty easy to attach to the inside of a cabinet.

-Paul

Yeah this is the standard shape for frossover inductors. I was curious why a torroidal shape wasn't used even if it wasn't the most efficient in wire usage. I'm more looking at the coupling affect of the fringe flux field. between two aircore inductors in the same enclosre.

The advanced speaker designers are shorting the inputs on the tweeter circuits and driving the midwoofer circuits to full power and measureing the unduced voltage at the tweeter terminals. This is mostly from inductive coupling. The crossover components are moved around unitl this inducted voltage is minimized hopefully to the limits of measurement. This is somewhat difficult to do with a smallish enclosure that speakers are designed in today.

IF there were an air core inductor design that was self shielding. That would be the bees knees for small speaker builders.

On Wed, 4 Jun 2008 06:17:33 -0500, "Mook Johnson" wrote:

....snip!.......

If you did "self shield" an air core inductore, you would do it by cancellation of fields, and that would reduce the inductance. The "air core torroid" that was mentioned in an earlier post would be very ineffective as an inductor, since there would be a large degree of cancellation. To reduce cross talk, the coils must be spaced away from each other where the fields are more unform, and then "orthogonally" positioned so that they are at right angles to the others magnetic field. Any ferromagnetic objects nearby will distort the magnetic field, and reduce the cancellation. Naturally, you could put a core around the whole thing, and realign the fields to stay within the core because of its much lower reluctance. That would be expensive, and a real nuisance. I can't think that the crosstalk is all that much of a problem.... very little power would be coupled from coil to coil, unless they are positioned one on top of the other. So what if you find a few milliwatts of low frequency at the tweeter? I would think because of physical proximity to the woofer, that the acoustical crosstalk is many orders of magnitude higher! Crossovers are pretty linear devices, and unless speakers are highly overdriven, they are linear too. So if there was any crosstalk (or infiltration of frequencies on the skirts of the crossover filters), there should be little or no audible effects. (Oops.... we're getting close to Bi-Wiring issues here) If the devices WERE nonlinear, you would have an issue, due to modulation effects (IM - Inter Modulation distortion). Why the concern about attenuated "out of band" signals getting to the speakers? Is there something I'm missing here?

-Paul

Mook Johnson a écrit :

You don't need to have all the inductors in toroid form (reciprocity).

--
Thanks,
Fred.

NMR systems use self-shielding solenoids, both for the main superconductive magnet and for the pulsed gradient coils. The structure is pretty much a compact solenoid on the inside of a bigger, looser solenoid, driven out of phase.

They have reasons. But for audio, this is just more of the standard insanity.

John

I had a crazy idea the other day - a solenoid has a toroidal magnetic field, and a toriod has a "solenoidal" magnetic field - if you swapped E with H in the formulas, would they still work?

Thanks, Rich